por
John R. Fischer, Senior Reporter | March 19, 2019
Glassbeam's Clinical Engineering
Analytics (CLEAN) blueprint
Baylor Scott & White Health, the largest healthcare integrated delivery network in Texas, will soon have the backing of machine learning and AI for predicting downtime and improving uptime of its imaging solutions, courtesy of a new partnership with machine data analytics provider Glassbeam.
The national top-tier IDN will utilize the vendor’s Clinical Engineering Analytics (CLEAN) blueprint to improve machine uptime and convert unplanned downtime into planned maintenance windows of the growing fleet of imaging and biomedical equipment under its clinical engineering group. Included in the installation will be the integrated hospital equipment management system of computerized maintenance management systems vendor EQ2.
“With a diverse set of imaging fleet from multiple manufacturers like GE, Siemens, and Phillips, it is critical to have a single pane of glass to manage and predict alerts and failures for BSWH’s field workforce,” Puneet Pandit, CEO of Glassbeam, told HCB News. “Such a solution is only possible with a centralized data platform that normalizes all data and powers alerts and predictions based on expert rules and AI/ML algorithms.”
Ad Statistics
Times Displayed: 12513
Times Visited: 21 Final days to save an extra 10% on Imaging, Ultrasound, and Biomed parts web prices.* Unlimited use now through September 30 with code AANIV10 (*certain restrictions apply)
The CLEAN blueprint is the industry’s first multi-modality and multi-manufacturer IoT analytics solution, according to Glassbeam, and is expected to improve medical equipment uptime at BSWH by more than 99 percent, surpassing the industry standard of 96 to 97 percent.
Its installation reflects the growing interest in IoT technology in which connected machines are expected to power “Smart CMMS” software like that of EQ2. Such solutions act as a ticket management system that tracks issues, such as downtime, in a variety of devices, from syringe pumps to MR, CT and ultrasound systems.
Their use, however, is limited by a reliance on human-made data entry, an issue which EQ2 has sought to rectify by
establishing a partnership with Glassbeam this past November at RSNA in Chicago.
Under the collaboration, Glassbeam’s machine learning capabilities will be integrated with EQ2’s Hospital Equipment Mangement System to validate the performance and log data of the machine, saving time and verifying the accuracy of findings faster.
“Instead of human intervention to create service tickets based on reactive calls from facility administrators, this joint solution will tap into machine data that will call home on a periodic basis (about every 15 minutes), act on rules, and open service tickets to make Baylor Health's service staff more proactive and predictive,” said Pandit. “In addition, integrating all service ticket data with the Glassbeam platform will create one single dashboard to track real time service metrics like mean time to resolution (MTTR) and mean time between failures (MTBF), among others.”
The basis of the BSWH-Glassbeam partnership is that combining machine logs with DICOM and HL7 data will drive dashboards in helping to acquire deep insights on utilization at machine and facility levels, not just for clinical engineering, but for radiology groups too, in the future.
The addition of Glassbeam’s software will also enable BSWH to build an expert system knowledge base of machine and part failures, and apply AI and machine learning to search error patterns, allowing it to form new rules for how to take proactive action when signs indicating the risk of unplanned downtimes appear.
The first phase of the installation will oversee the connecting of 90 CT and MR systems across 22 BSWH hospitals in Texas, after which it will expand to include other modalities and locations as the solution matures across a subset of facilities, creating a pervasive internet of medical of things application for BSWH.